A comprehensive survey on Carrier Ethernet Congestion Management mechanism

Abstract

In the literature, congestion control schemes have been broadly studied. Nevertheless, this subject needs to be reviewed in the context of Carrier Ethernet. Congestion happens when the capacity cannot cover the resource demands. Nowadays, users are witnessing the networks saturation because of the expanding demands for bandwidth. The addressed issue is not only how to avoid congestion but also how to use all available capacity without overuse it or underuse it. To deal with congestion, link level algorithms drop or mark packets with increasing probability as buffer congestion increases. The most common solution is based on the transport layer algorithms to adjust the resources transmission rate by using these dropped or marked packets. Carrier Ethernet defined by layer two uses new proposals specified by the 802.1Qau standards committee in order to handle congestion problems. This paper surveys and studies different properties of Carrier Ethernet congestion control schemes. This paper also draws a parallel between the different schemes and point out the advantage and disadvantage of each one. Then, this paper presents a taxonomy of the Carrier Ethernet congestion control mechanism and correlate it with existing taxonomies.